1. science
2. social science
3. history
4. archaeology

!∀#∃∃%&∋&∋&∋
&(
(
(&)∗(
!∗!#%)(+,
−
+&
.&
∀/.&
(& (01
/
!0
!2!
3)4/5678∃9
:
Why that? Why there? Why then?
The Politics of Early Medieval Monumentality
Marfin Carver
,
b
The hypothesis presented in this paper has already appeared in various
fragmentary forms (Carver 1986,1993,1998b), but has not hitherto been drawn
together, so it is fitting that I should k y to do this in honour of my best teacher
and most telling critic. After all, it would not do for Professor Cramp's
Festschrift to be too burdened by flattering emulation; it should contain
something exasperating as well. So I look forward to her leafing through these
pages with growing despair, culminating in the tart response familiar from my
carefree student days: 'really Mr Carver, I have no idea what you are talking
about', meaning: 'actually, 1 know perfectly welI; you on the other hand ...'
In brief, the hypothesis concerns the history of the early medieval period
in north-west Europe and our ability to read it from archaeology, or more
specifically from its major inves.tments such as burial mounds, churches,
illuminated manuscripts and sculpture: the word 'monumentality' of my title
is intended as shbrthand for aII these things. Confidence in the idea that
monuments had (and have) a meaning beyond some vague celebration of an
individual or propitiation of an unseen omnipotence has been growing among
prehistorians (e.g. BradIey 1993), and is an accepted feature of the historic
period. We know that monuments are more than passive memorials because
written commentaries, poetry and inscriptions declare their active purposes for
us. Monuments comprise the vocabulary of a politicaI language, fossilized
versions of arguments that were continuous and may Rave related more to
.what was desired than what had occurred (Carver 1993).At the same time, we
need not suppose that the expression is necessarily so subtle, sceptical or to use
a fashionable word, ironical, as to lose all hope of making equations between
a socieiy and its ideas. That architecture, sculpture, burial mounds and
brooches have messages beyond the functional which are dependant on their
social, economic and above a11 their ideological context was never an issue: to
understand their real meaning is the goal and the aim of each generation that
\
Martin Carvw
studies them. It is very likely that the motives .I attach to the construction of
the monuments to be discussed are equally inadequate characterizations of the
profound stresses that motivated and were concealed by their makers. That
said, the monuments are what survive, and our story must temporarily keep
the candle burning untiI their story can be told.
Some assxlmptions
The argument requires a number of assumptions to be declared. First, all
archaeologists are obliged to acknowledge that whereas their efforts have the
advantage of generating new evidence, there is never a good moment to argue
from it; even newer evidence can quickly take the shine off too detaiIed a
model. Reasoning from documents has the disadvantage that so few survive,
but the advantage that no new sources are likely to appear suddenly in the
midst of the six-year-long composition of a major synthesis. Archaeological
reasoning and historical reasoning have naturally different rhythms, and the
world is a more interesting place because of it. But, in order to offer a
historically acceptable model, the archaeologist is oLliged to assume that it is
legitimate to argue from the material culture we have; to assert that in certain
matters, such as the occurrence of churches and burial mounds, the
distribution will not now alter markedly: that what we have is not exactly
what there was, but is an acceptable representation of what there was. Having
declared this assumption, the archaeologist should be permitted to develop a
model on evidence that is partial; the documentary evidence is partial too, but
it is legitimate to afdempt to write history from it.
A second assumption is that the repertoirgbf material culture at our
disposal is heterogeneous and cannot be interpreted through a single
theoretical exegesis. Not every object or site has an equal claim to be treated
as intentionally expressive. If the Sutton Hoo mound 1 burial or the
Lindisfarne Gospels can be seen as having agency, representing material
culture in its active voice, there is no need to impute the same intentions to a
spade or a spindle whorl. The old definition of a 'culture', pulled tlus way and
that since Gordon ChiIde used it the preface to The Danube in Prehistory (19291,
can be seen as neither all cultural, nor always seeking identity, status or
affiliation, but multi-purposed, a set of different statements addressing
different audiences or none. If economic information is incorporated in the
layers of midden heap, political meaning is most likely to be embedded in sites
and objects of high investment and public access. In our period, the prominent
candidates are burial mounds, jewellery, churches, illuminated manuscripts
and sculpture. These appear not only in a single place at successive times, but
at the same time in different places. If assumption number 1 is tenable this
variation has a meaning and was intended to have one.
If the choice of the vehicle of expression is intentional, how far is it sui
generis and how far is it owed to the emulation of the neighbours? This
'
The Politics of Early Medieval Mdnumenfality
3
depends firstIy on whether the neighbours are visible, and the third
assumption I make is that in early medieval, northern Europe they were. The
Anglo-Saxons knew about the Romans, the Irish, the Picts, the British, the
Franks, the Swedes, the Danes and the Norwegians at both a general and a
personal level, and we can infer this both from books (Bede) and from graves
(see, for example, Hines 1984). It is these contacts and the transmissions
between them that allow us to suppose that the commissioners of seventhcentury monuments could fish in a large reservoir of ideas. George Henderson
invokes the broad range of stimuli available to the composers of the Book of
Durrow in AD 680: 'Discrete national traits, in design, techruque and the
motifs, came into conjunction in the peculiar, small-scale, packed
selecba,of
circumstances of the British Lsles, with its ever-changmg political scene, under
the constant cohesive influence of the internationally active Christian church,
writing and speaking Latin to all ears' enderso son 1999,53). But .to these same
people, the local traits of Scandinavia an$ Saxony were also visible or at least
were known - the patterns on cremation urns, the wrist clasps and squareheaded brooches which had been worn by East AngIian nobles. The bracteates
or guldgubbe made in Fyn are sources for Insular Art potentially no more
obscure than Coptic bowls or stamped Mediterranean pottery. So why were
their ideas not recycled on the pages of the gospel books? The animals of the
Sutton Hoo shoulder clasps and purse lid and of the Durrow carpet pages are
links in the same technician's chain. But a great deal more of the Sutton Hoo
menagerie, on helmet, sceptre, and sword was not adopted by the Christian
artisans. Eastern Christian figurative art was known to the kings of East
Anglia, as can be seen from a sixth-century bucket found in a field a few
hundred metres north of Sutton Hoo (Mango et al. 1989). But it was not
incorporated into the Sutton Hoo jewelIery. The point is an obvious one. The
Anglo-Saxons were not 'eclectic' in the sense of taking motifs at random from
some workshop floor or exotic street scene. They were not indiscriminately
'influenced' by things they had seen in the halls of Danish relatives or on trips
to Rome. They were creative and seIective, not eclectic, but choosy. There was,
potentially, a broad range of accessible options and since the different options
were equally available, the choice which was actually made must have a
meaning.
The next assumption is that the repertoire of possible choices does not
need to be contemporary; in other words the maker of expressive artefacts and
monuments can also fish in the pool of the past. The metal-smiths of later
Celtic Britain and IreIand can revive La T h e styles after an interval of 500
years, just as Roman motifs, lettering, ornament and pottery are reintroduced
in seventh-century Northumbria, the courts of Offa and Charlemagne, the burh
of Alfred, and in numerous subsequent renaissances (Carver 1993). It was
recently argued that the notched shield seen on the St Andrews Sarcophagus
derived from something last seen (so far as we are aware) in the Iron Age of
southern Britain (Carver 1999a). There are two ways in which this transmission
4
Martin Carver
can be achieved, the most easily acknowledged being swvival of the obje~tor
site itself. Bailey (1992) suggests a 'co,nseflatism' to account for the gap
between Sutton Hoo a n d Durrow (see below), while Henderson prefers to see
the pqsrnjssion occurring via suryiyingpat&rn books (Henderson 1999,SO).
Roman things no doubt turned up or were dug up from time to time, certainly
when former Roman sites were being-redeveloped f r ~ m
the- t~-th-century
pnwards; this easily explains why the form of late Saxon pottery was drawn
from Roman, not Frankish models (Carver 1999b, 42). The argument has been
extended to sites; former prehistoric ceremonial centres being commandeered
by later authorities for their own legitimation (Bradley 19881, and it is likely
that the Anglo-Saxons had quite a sophisticated knowledge of 'landscape
archaeology'. Bronze Age burial mounds were suitable places for the emulation
of Pagan status, wliile Christian missionaries should be assigned old Roman
forts like Burgh Castle (Johnson1988).
It is more difficult to use the 'swvival' argument to explain the readoption
of certain other practices such as ship burial, arguably seen previously in that
form neither in Scandinavia nor Britain when it was practised at seventhcentury Sufdon Hoo and Snape. Here the assumpiion required is that the idea
of ship-burial was present in the previous century and perhaps long before,
but not thm practised. The archaeologist cannot excavate these ethereal
images, which are transmitted through unrecorded space like cluldren's rituals
in the playground (Opie and Opie 1977). Unless we suppose an early
archaeological expedition to the Pyramid city at Gizeh, the idea of shp-burial
came out of the heads of people and became archaeologically visible only
when it was reified in some moment of exceptional stress. Ship-burial is
therefore not a custom but a statement in context, and our interest in it isnot
so much in the innate meaning of the ritual as the meaning it had for seventhcentury East 'Anglia: why that, why there,. why then? (Carver-1995),
The contextual explanation may be thought to deserve precedence over the
cultural. For example, there were burial mounds in the fifth?sixth, and seventh
centuries in Britain, but this is not simply a 'burial custom' that evolves or
endures. The style and location of seventh-century mounds is different from
*heir Anglo-Saxon predecessors, being larger and more solitary (Shephard
1979). In any case, the burial mound does not have to carry exactly the same
significance in the seventh century as in the sixth: it depends what else was
happening. At Sutton HOOthe use of burial mounds can be deemed
'expressive' or even ~ o l e k c aon
l two counts: first they represent a highex:level
of investment than the mounds of previous centuries, and second, they are
constructed a t a time at which quite different monuments are being
constructed in adjacent lands, for example in Kent. By contrast, Kent, rich in
burial mounds in the sixth century, is investing in a new repertobe in the
seventh - the apsidal churches.of Canterbury.
If this is merely to say that mon~lmentshave political meaning, and the
choice of monument reflects a political agenda, that is already a great deal for
The Politics of Early Medieval Monumentality
5
some to accept. But it is auo-qatic for what follows. I want to propose that
early medieval investments, prestige buildings and artefacts, can be used to
write political as well as cultural lustory, and that if this is acceptable, we can
paint a picture of confronted
-- poiities pursuing different ag~ndasduring the
f i f h to the eighth centuries.Eoth iviilun the %and 'of ~ r i t a i nand beyond.
Fig. 1.1. Monumental mound-burial in Briiain in the sixth and sevenih
cenfxries AD (Carver 1986, after Shephard 1979).
\
Reacfionay mound-building
The conflated version of Shephard's map (Fig. 1.1;Carver 1986) shows a trend:
cemeteries with numerous small burial mounds are present in sixth-century
Kent, and larger more solitary barrows are built in the late sixth into the
seventh century outside Kent. We should seek some explanation for this
Fig. 1.2. Bihrial under large mounds, sixth t o lak seventh century AD in
the Rhineland {Bohme 2993, Abb. 101).
apparent contrast, and can easily find one in the pivotal ideologicagevent of
the period - the documented arrival of the Christian mission from Rome in
597. If Christianity depressed1barrow-budding in Kent, it provoked it among
the umonverted neighbours in a more militant form. Taplow, Sutton Hoo, and
many other monumental constructio~of around 600 in Saxon lands, can be
read as reactions against the Christian mission (Carver 1986, k998a, 1998b).
The work of H. W. Bohme enables us to propose a similar process for the
~hineland(Bijhme 1993; Miiller-Wille 19981, @re the building of monumental
mounds begins at the ~ n mouth
e
in the fifth century and moves s t e a a y
upstream, arriving in Switzerland by the eighth (Fig. 1.2). In an equally
thorough census, Bohme has also tracked the building of early chwchesiq the
same area; this second lund of monument can be seen to sbadoy the first - the
earliest
churches replacing the earliest barrows and the latest chwches
.
following the later barrows (Fig. 1.3). This should mean not just that churches
replaced monumental barrows: they provoked their construction in the first
place. We can imagine h s 'bow-wave' effect taking many forms. When the
' mission of Adam of Bremen arrived at the ceremonial cenbe of Gamla Uppsala
in Sweden, the grotesque sacrificiaI practices were noted: apimals and men
were hung on a great tree - and it was naturally assumed that this was a first
encounter with an annual barbaric custom. Such it may have been, but the
practice of sacrifice has been hard to find archaeologically, except on a small
The Politics of Early Medieval Monumentality
7
Fig. 2.3. Burial in or in associafion wifh churches, sixth to l~feseventh
century, in the Rhinellznd (Buhme 1993, Abb. 99).
,
scale - for example animals slaughtered and included in graves. Possibly the
text is intended to exaggerate the barbarity and justify the incursion. But if the
event really took place, it might have done so out of the insecurity and anxiety
caused by the Christian presence itself. The sacrificial practices were not so
much discovered by the Christians as provoked by them. The same might have
been true of Cortks in Mexico; the increasing orgy of Aztec sacrifice being
caused by fear of'the alien adventurers who denounced it.
Leslek Pawel Slupecki (1997) argues for similar processes in a study of the
impact of Christianity on the eastern Slavs, and cites earlier work along the
.same lines (for example H. Jowmianski in 1979): the observed surge in Pagan
religion was an ideologcal response to the advances of Christianity. Human
sacrifice increases, temples are built and the role of the priest develops. There
was even an example of altars to pagan and Christian gods being placed in the
same temple, as had been notoriously perpetrated by the East Anglian king
Raedwald in similar circumstances ,WOgntururi earIier and a few thousand
rnilesxestwards (STupecki 1997,186). The p r o x i m i ~ o f~hristi&f~rovoked
opposition to its ritual power, a menace occasionally molified by dressing in
its clothes.
It would not do to ascribe every monumental burial mound in northwest
Europe to anti-Christian anxiety; but it must have played an important role.
The general map produced by ~ o b e rv&
t de Noort (1993; now amended Lutovsky
Martin Carver
Fig, 1.4. Distribution of dated burial mounds and burrow cemeteries
(including some graves with ring ditches, thought to have been mounds)
(Lufovssky 1996, fig I b).
1996) shows a chronoIogica1 trend which for the most part pre-echoes
conversion (Fig. 1.4),but not every case fits the model. The mounds at Hogom
in Medelpad, Sweden (Ramquist 1992) are too early (at fifth or sixth cenfmy)
to be reacting to a Christian menace. But that is not to say that no equally
stressful circumstances were involved; for every monument we need to seek
both the impetus and the motive.
The Politics of Early Medieval Monumentali~
9
This argument suggests that the building of a monumental burial-mound
is best regarded as a historical event, the result of the conjuncture between a
set of beliefs and the circumstances of the time. The circumstance most likely
to provoke these monuments is political insecurity, and Christianity with its
declared programme of conversion and Roman imperial associations must be
counted among the provocative forces of the age. Is it right to regard the
signals of barrow-burial as 'poIitical" (in that they are seen as mainry
expressing power, alignment and allegiance), rather than ethnic, customary or
religious? They could be seen as ethnic in the sense that a certain group of
people who regard themselves as related might act togekher; or customary, in
that barrow burial had existed previously amongst those people as a smallscale ritual ready for political inflation when needed; or religious in the sense
that a common set of beliefs might be implied in order fox similar events to
provoke a similar response. But these factors are neither necessary nor
sufficient for monumentaIity, and they do not explain why monumentality is
intermittent. Gqjy .politics-are sensitive to events, so politics should Iie behind
change. If political motivation has primacy, ths would help to explain why
pedpl+ Iiving in different lands can use the same
-- -s i ~ a lh-emulation
s
of each
other, without c l a i m 3 ethnic affiliation or acknowledging the same gods. So
to
Sutton Hoo, forexample, we do not need to claim that East Angles
worshipped,, Odin or were Swedes; only that they well understood the
philosophies of Pagan thinking and the language of its monumentality. In the
sixth century the East Angles muttered their allegiance to the ideoIogy, but. in
the seventh they felt obliged to shout it.
-
-L
I
l
m
A monumsntal hiatus
Sutton Hoo ceased to be used as a high-status burial ground about AD 625,
and became for the next 200 to 300 years a place of execution: the ideological
struggle had been' resolved in favour of ~igorous
_ _ Christian
--- kingship
.
(Carver
1998a). In &g- aftermath to the documented conv&iion, however, and not just
in East Anglia, there is a curious lull in material expression of nearly half a
century from about 625 to-675, when soaety seems to be holding its breath.
Furnished graves have proved difficult to assign to this period (Geake 1997,
124), as have churches (Taylor and Taylor 1965), sculpture (e.g. Hawkes 1999,
404) and illuminated manuscripts (Alexander 1978). It is has to be assumed,
nevertheless, that churches were being built and monasteriesfounded, because
the documentary record tells us so. T h e sixtl-century monasteries of Iona and
Wluthorn were active in the north and the episcopal centre at Canterbury
active in the south. A church of wood was built at _Yockjn 626 and a bigger
one planned in stone. Aidan w a s in Lindisfarne and Birinus at Dorchester in
635. Ceanwealh built the church of St Peter at Winchester in 648. Christianity
became compulsory in Kent in 640; temples were destroyed in Essex in 665.
Barrows may also have been built: Fenda kept the pagan flag flying until his
_--
+
10
Martin Carver
death at The Winwaed in 655. But investment in objects and constructions on
the monumental scale are difficult to find on the ground, and rnaybg ma.tgria1
expression was deliberately modest. Perhaps this was a period in which the
actual message of Christian purism was briefly effective, and even the
aristocracy lived briefly by precepts of generosity towards others and reserve
in celebrating themselves.
On the boundaries, as we have seen from the case of the Rhineland, a
more Iively monumental confrontation may take place. A British exampIe
seems to have been discovered by Edwina Proudfoot in the course of her work
at the Hallow Hill and amplified in a brilliant paper by the late Ian Smith
.(Proudfoot 1996; Smith 1996). Here a range of different monument types square-ditched barrows and Class I symbol stones in the north and long cisr
graves and stone crosses. b the south are confronted across a seventh-century
boundary running across Fife. Objectionshave been raised that long-cist burials
do not have to be diagnostic of Christianity, because they are found in
prehistoric variants and are distributed further north. But they certainly
resemble the Christian cist-graves of the Alpine region (Carver-1987) more
closely than any Bronze Age predecessor. If it were an attribute of Christianity,
the long-cist would be found to move over Scotland revealing the itinerary of
successful conversion. The early Christian pressure on the Picts intimated by
Bede (HE U1.4)is persuasively captured in Smith's snapshot of a monumental
frontier (Fig. 1.5)
The Northurnb r i m forum
Further south, the monumental machine seems to start up again at the end of
the seventh-century. Ripon an$ Monkwearmouth are founded, Brixworth built,
the Book of Durrow and Durham A 11 10 illuminated. It is of the greatest
interest that, inNorthumbria and the other English kingdoms, furnished burial
also restarts in the later seventh century and continues into the early eighth
(Geake 1997, 124). It is furnishing of a very particular kind: necklaces and
ornaments of a marked Roman and Byzantine (rather than Frankish) character
are found predominantly in female graves. One burial with a pursefd of
sceattas at Garton-on-the-WoIds, is certainly as late as 720. By this time, plain
incised grave-markers were being erected at Whitby and Hartlepool, the
' - Lindisfarne Gospels with its virtuoso insuIar ornament had been completed at
Lindisfarne, at Jarrow Bede was writing his history and the immense Codex
Amiatinus with its meticulous adherence to Cassiodorus' classical prototrpe
had been lost on its way to Rome. Northumbria in the early eighth century
was not only a hive of creative endeavour; it was also a mass of alternative
expressions, the re-ified ideas of clerics and warriors, northerners and
southerners, men and women, those who live 011 the Wolds and those who
lived in the Vale of York or by the Tees, the Wear or the Tyne. The society W-%
presumably stable enough to allow differing opinions and the distribution of
The Politics of Early Medieval Monumentality
1Class I stonels
't Eccles place-name
O long-'cist cemetery
if
R
square-barrow cemetery
St Ninians (Ecciesj
Fig. 1:5. The seventh-cenfuryargument in Fye. Map prepared by Inn Smikh
(Smith 1996, fig 2.5).
resources was generous enough to allow plural and varied expressionsin stone
and vellum.
In 710, Nechtan, King of the Picts, sent to Ceolfrid, abbot of Wearmouth/
Jarrow for advice on t h e expression of Christianity in the Northumbrian
manner: not just how to calculate the date of Easter, but the correct form of the
tonsure, and how to build a 'Roman' church (HE V.21). And yet the
consequences of his initiative seem to have resulted in no network of monastic
scripforin producing Pictish gospels and hagiographies. What they have left us
is something very different: scores of monumental slabs, "carrying Pictish
12
Marfin Carver .
syg-tb~lsas well as Chrishnmatifs that were famiiliar in the local repertoire
of Iona or mr&umbria.
It is hard to resist the conclusion that Canterbury, Jarrow, Iona, the Garton
burials, the Codex Amiafinus, the Lindisfarne Gospels, and the Class I1 Pictish
stones on Tayside actuallyrepresent different reactions to Christianization. Was
this because the peoples concerned were ethnically distinguished, or felt they
re? Angles, Britons, Scots and Picts spoke a different langiiage, Bede tells us,
so why should the language of their art and material investment not differ too?
Yet it seems stTange that a shared ideology should take so many different
forms between neighbours, and in the case of the Garton burials and the Codex
Amiatinus, the same linguistic group was presumably involved. Do the late
seventh- and early eighth-century burials proclaim an enclave of dissident
women, resentful of the Chstian project? Or are some women gqardbgtheir
family claims by imitating the imperial ornaments of Theodora, in order to
express aristocratic membership of a Roman 'folki, as opposed to the illiberal
clerical alliances of wjl&id?
On the supposition that material culture is meaningful, and that the style
of investment is deliberately chosen, we are entitled, I think, to read into it
differen~po&ic.a~s_tances.
There seems no good reason to allow these forrns of
expression to be simply local, cultural, vague, ill-informed, confused or
'syncretic' (Fletcher 1997, 126-153). That seems a negative way to view a
people who emerge from the pages of Bede as predatory, extravagant and
fiercely opinionated, just as IikeJy to express themseIves forcefully in words
and material culture as any other privileged class through the ages. Conversion
did not, it seems, produce a homogeneous culture for us to find, but appears
to have expressed itself in very different ways - differences here ascribed to
politics. How could such political stances be defined? The archaeoIogy has a
pattern to it, and the documents give us hints of inter-Clwistian dissent, so as
a first step we could set out to seek correlations between them.
DzfjCerent roods fo Christianify
Since we are concerned here with political signds within communities that
have adopted Christianity in some form, we should be searching for a model
that distinguishes between the different ekindsof Chstian organization and
economics, ~t least three varieties of Christian organization or socio-economic
control are 'well known and provide us with a starting point. In the episcopal
system, a bishop holds authority over a territory (the diocese) and draws
revenue from it. This is the system that most closely echoes Roman imperial
administration and requires a similar method of taxation to run it. The
archaeological correIate is the episcopal centre of which the finest example has
been given by Chades Bonnet's excavation of the episcopd group at Geneva
(1993). The attributes are the basilica, itself a type of Roman government
building, the baptistery and the pulpit. At Geneva, the baptistery, initially a
.
The Politics of Early Medieval Monumentality
43
small tank, developed with the increasing demand until it acquired its own
supply of piped water. The trappings of an episcopd church, with its imperial
aura, can be contrasted with a materially more elusive but intellectually more
influential variant, the monastic church. Whatever the detailed circumstances
of early monasticism, it has the genera1 appearance of an autonomous
movement, if not exactly dissident then at least less a creature of state than the
hierarchy of bishops (Chitty 1977). whatever else drove the earliest eremitic
monks to the desert, an ec~nomicconsequence would be a freedom from tax
or m e . When the hermits clubbed together in the coenobitic communities, the
. seeking of independent non-taxabIe endowment would have been an early
method of survival, one which was later adopted in northern Europe. Colmba
received Iona from the Picts and King Oswald 'gave lands and endowments
to establish monasteries' including Lindisfarne (HE III.3) and Jarrow was
supported by lands on which Bede was born (HE, see autobjographical note,
p 336).
The Christianity practised by the monasteries of Egypt wouId not have
been organized and delivered in t h e same way as the imperial Christianity i t
Geneva or Hagia Sophia, and it seems highly likely that there was
confrontation. Such a confrontation has been postulated as endemic in the late
Roman town, where the 'episcopal group' is situated inside the official city and
grafted into the late Roman administration, while the 'monastic centre' grows
up in the extra-mural cemetery around the person of a martyr (Perinetti 1989),
Either or both these cenfxes may then act as a kernel for the growth of the
early medieval town, providing two nucleii which each presumably had its
declared and fiercely argued rationale. This phenonemon, which Perinetti
t e r n "bipolarity', can be seen at Aosta and Salona. At Geneva, the episcopal
centre within the walls wins the battle for hearts and minds and town lay-out,
while at Tours the monastery proved to have the greater magnetic force.
The archaeological correlates for northern monasteries earlier than the St
Gall plan in the 'ninth century are notoriously hard to define. The remains in
the Egyptian desert invite us to look for simple cells of hermits, perhaps
clustered together, a vision which has influenced the identification of Skellig
Michel and Tintagel as monasteries. There has long been an assumption that
early insular monasteries should be sited in an oval or circular or curviIinear
-.. - - the British Iun of Wales and Cornwall. Few of these have been
enclosure,
dated: at Whithorn, a ditch 10m Iong and lrn wide containing an undated
sherd of glass was extrapolated into a monastic boundary enclosing half a
hectare (Hill 1997,29,77). At Hoddom, the vallum and pallisade, traced on the
ground as a D-shaped enclosure, gave five radiocarbon dates in the seventh
century, prompting the excavator to suggest a date of construction between 600
and 680 (Lowe, in press). The sites of some documented Irish monasteries (for
n j found in association wkk circular bank~d
example that at ~ i l t i ~ m a are
enclosures (Mytum 1992,80); but in so far as they resembl; the prehistoric rdh,
they might represent the recycling of a local idea (Ryan and Mitchell 1998,
14
Martin Carver
260), while other forms of monastic enclosure, such as that at ~lonma-ise,
are apparently rectangular. The site of the present monastery, on Iona is
enclosed by a penannular bank, but this has lately yielded a prehistoric
radiocarbon date (Fisher 1996). The possibility must be considered that
monastic communities took whatever disused fort they were offered by the
king, as did Fursa when he moved into = h eredundant: Saxon Shore fort at
Burgh Castle, courtesy of King Sigebert of East Anglia (abbe).
.
h i d e the enclosures, we look for ways of distinguishing the monastic
from any other kind of estate centre in which there has been educated
investment at a high level. In the north, a number of the documented monastic
sites are signed by grave markers with simple crosses and inscriptions: for
example Jarrow, Monkwearmouth and Hartlepool, Whitby has 41 (Hawkes
1999) and Iona has numerous examples (Fisher 1996), although many of these
may have accrued during the lingering aftermath of the early Christian period.
Monasteries use and produce books, so that vellum p s ~ d ~ c or
~ ofinds
n of styli
ought to be d i a ~ o s t i c ,But accepting thdi.-"the middle Saxon aristocracy
provided leadershp both for those who fight and those who pray, it would not
be impossible for attributes of the literate sector to appear on a seigneurial
estate (see Loveluck, this volume). As we can so far see them, the markers of
monastic centres, while they kcho thos;! of a secular estate, at least differ from
those of episcopal centres. Since the peoples of Britain and.Ireland potentially
had access to the full range of Christian models, we are entitled to deduce that
such a difference had a meaning behind it and was deliberately chosen.
A people unused to taxation would find the endowment of monasteries
preferabIe to the imposition of a permanent revenue required to support
bishops. But another tendency has long been present in Christian practice, and
there seems good reason for supposing that it would have been an option as
early as the conversion period; it would moreover require far less of an
organizational upheaval in a pagan people. In this adoption of the Christian
structure, the faith is administered neither ,by bishops or monks, but directly
by the landowner or local aristocrat who can appoint and pay his own priest.
Such a 'secular' or 'private' option should be. signdled by individud a$
decentralized investment, so is bound to be harder to see. The most persuasive
examples are provided by those communities that erect monuments in
dispersed patterns - patterns which prompt an association with estates. The
$pression of se'cdarity is reinforced if the iconography is varied and features
everyday life, rather than being merely iconographic, repetitive add orthodox.
The Pictish Class II and the Anglo-Scandinavian monuments seem to be strong
candidates in this respect (Carver 1998~).
A less visible but equally telling example is offered by the small personal
funerar'y crosses found in central Europe, and recently studied by Miiller-Wille
(1998b; Fig 1.6). In fifth-to seventh-century Trentino, cross-shaped brooches
were deposited with females; in the central Rhine/Mosel area, cross-shaped
brooches of the late seventh century were also found in the graves of females;
'
The Politics elf Early Medieual Mon~amentality
Fig. 1.6. Distribution of sixth- l.o eighth-centek y gold foil crosses in the
Alamlrnnic urea between the Rkine and Danube (Muller-Wille
1998b,,fig 6).
while in the Alamannic district between M e and Danube, both sexes were
buried with gold foil crosses from the sixth to the eighth century. In each case,
the crdss was the main Christian signal. It is specificdly noted that in
Alamannia none; of the gold-foil crosses was found in connection with the
earliest churches; moreover this is not an imitation of an earlier Christian
practice, since no grave with a cross is known from the period from the first
to the fourth century in t$e Mediterranean. One reasonable interpretation of
this pattern is that the date and distribution of the burials accurately reflect the
progress of a gradual conversion northwards (ibid., fig 3); but without having
to challenge that general picture, it can be noted at least that the acceptance qf
Christianity was signalled in these territories in different ways. If the converted
landlords of the RhineIand repIaced their barrows with churches, and the
citizens of Geneva expanded and improved their massive episcopal centre, the
people of Namannia, sandwiched between them, induIged in the less
extravagant practice ofsewing small gold crosses on to the tunics of their dead.
Was h s modesty, poverty or ignorance? Or does the chosen signal represent
a non-centralized economy and a belief in maintaining authority at community
level - in short a different political stance?
16
,
M ~ r f i Carver
n
This case for a contiontation between Christian prescriptions depends on
the archaeological identification of their correlates and their distribution. But
some support can also be found in the sparse documentation, for'exam~leas
was revealed in Wilheh Levison's cIassic essay of 1946, which records
instances of tension between bishop and monastery and Christianized
aristocrats in the eighth century. Already in 672 the Hertford Canons state 'No
Bishop is to disturb a monastery or take its property by force' the implication
being thqt they might be tempted to do so {Levison 1946,231, while the Synod
of Cloveshoe in 747 was aimed at restoring 'the hierarchical order and a
regular church administration - provinces under an archbishop in direct
contact with Rome, dioceses with a monarckucal bishop supervised by the
archbishop and controlling the whole clergy of his district' (ibid. 86). The
Bishop was sacramentally necessary but their function was sometimes
exercised by 'wandering bishops and other bishops without dioceses but
attached to monasteries' (ibid. 97). Politically powerful converts preferring the
,secular option could apparently operate in a similar way. Cuthbert md seven
continental bishops of English origin sent a letter to King Ethelbald of Mercia
to admonish him and reprove him for his dissoIute life and his encroachments
on possessions and privileges of monasteries (ibid. 92) - a glimpse of an
alternative secuIar church which probably delivered more satisfactorily the
economic and politicd interests of the Mercian aristocracy. Levison
summarized the tendency of landowners to form their own 'proprietkry'
churches and monasteries (Eigenkirchen, Eigenklosfer) and reminds us that 'such
tendencies might be a source of danger to the monastic ideals' (ibid. 28). It is
clear that bishop, abbot and Iord had a discourse in which much was claimed
and definitions were blurred. It would have been in the interests of Bishops to
claim jurisdiction although they had none, and for landlords to acknowledge
Rome but to pay nothing and do nothing which would encourage Roman
control of their affairs. Wandering bishops would have been useful in
ordaining priests and giving them liturgical authoriiy and sacred powers, but
they would have brought no necessary allegiance to an archbishop or to the
whole Roman apparatus - of which the northern Europeans were (and still are)
endemically suspicious. We should expect ideological difference to be denied
in documents even while it was practised or proclaimed on the ground. This
suggests that the search for expressions of locd politicd resistance or
alignment through the study of monuments is a legitimate quest.
In Britain, perhaps the object of investment most sensitivd to the Iocal
Christian prescription is the standing stone monument, a field to which
Rosemary Cramp has made an immeasurable contribution. Lxke barrows, the
stone crosses may have started as memorials to an individual ancestor, whose
descendants continued to profess control of the same land. Like barrows, they
may have acted as places to meet and dispense justice. If family and power are
signalled in cross-slabs such as those from St Vigeans or Aberlemno roadside,
so are the elements of Christian doctrine; enough at least to merit the benefits
'
The Politics of Early Medieval Monumentality
17
of the Christian cornmercid network without sacrificing autonomy to either
king, bishop, abbot or pope. At a later date the cross might have been joined
by a church and this too might have been a local or private commission. If we
can make its socid label stick, the standing stone monument is going to be a
powerful tool with which to write history - because we have plenty of them
(see Edwards, this volume). A m a Ritchie (1995) has made the case that centres
such as Meigle produced monuments, especiaIly the widely distributed
standing cross-slabs, through lay patronage in the ninth and tenth centuries.
Such monuments, bearing a cross and other icons, were surely estate-markers
that reflected the taste and aspirations of the lahdlords, but also satisfied the
need for a ritual focus for a community accustomed to associate spiritual
orthodoxy with a local war-leader. George Henderson (1999, 209) observes:
'According to the Hodoeporicon or guide book, taken down from the verbal
report of St Willibdd, the English missionary who died as Bishop of Eichstadt
in Germany in 785, it was the custom on nobleman's estates to have the
stmdard of the holy cross set up as a focus for daily prayer ... The cross
monument was evidently a place of resort in time of crisis.'
The newly-converted Vikings seem to have preferred the secular mode of
Christian conh.01 to the monastic or episcopal model (Fig. 1.7). Jim Lang's
survey of East Yorkshire (1991) showed that the grave-markers of the seventh
and eighth centuries were distributed among a handful of monastic sites, a
point reinforced by Jane Hawkes (1999, 405) with reference to northern
Northumbria: of 'approximately 230 pieces of carved stone associated with preViking contexts north of the Tees, nearly 180 come from just five sites:
Lindisfarne, Hartlepool, Hexham, Wearmouth and Jarrow'. But this distribution
changes radically in the ninth century. In East Yorkshire, the monuments are
more numerous and more dispersed, as well as displaying a more individual
iconog;aphy. It was Rosemary Cramp's observation, made many years ago
(1980, 18), that 'the Vdcings secularised art just as they secularised
landholding', that set me thinking (in the context of the politics of our own
day) that perhaps the Scandinavian hostility was not to Christianity but
towards state control, taxation and the alienation of land to non-wealth
creators. The Vikings, by ,dissolving the monastic system and substifmting a
secular alternative, had succeeded in effecting a conversion of their own
(Carver 1998~).
Conclusion
'
These case studies, briefly expounded, are held to imply that the episcopal, the
monastic and the secular church represented three different strategies for
developing a Christian kingdom which has shed paganism. Each mode of
religious control had very different implications for the people - especially the
aristocracy - in terms of self-determination, inheritance, prosperity and
alliance. A fly on the wall d a Dark Age hall must have heard the argument
18
Martin Carver
over and over again: loss of sovereignty and increased taxation is the penalty,
heaven and an expanded market the reward. A soft strategy for the king (who
had most to gain from the church's protection of his dynasty) was to endow
some of his own land and encourage others to do the same, in the hope of
qualifyng for the Christian alliance, while avoiding the alienation of his own
peerage. But some Ieaders would not be strong enough to achieve even this
compromise. Landowners resist an irreversible commitment to extreme and
dirigiste doctrines, preferring to cherish the northernerJs ancestral ethos: every
man his own lord, in control of his own land and with his own link to God.
But there would be no harm in adding Christ's escutcheon to one's shield,
especiaIly if it earned useful contacts and warded off the threat of a take-over
or conquest on trumped-up ideological grounds.
HOWlikely is this picture? Was the eighth-century inhabitant of Britain
really capable of such ingenious decision-making? How far does this model
superimpose on early Christianity the intellectual fragmentation of alI that has
happened since that time? Does it urge a particular version of Christianity to
which certain local politics of our own day inclines? William Frend (1996, 61)
uses the term 'confessional interests' to describe schoIarship driven by a
particular branch of Christian ideology, showing how loyalty to the success of
the Roman imperial past had inhibited the archaeoIogica1 recognition and
rediscovery of the Donatists. But Donatism was a Christian variant which most
certainly existed, was professed, debated, opposed, championed and expressed
in monuments. The native inhabitants of Numidia (now Libya) had under
Rome adopted a religion ruled by a sacrifice-hungry god ('SaturnJ) which
demanded human victims, In 260-290 'the religion of the great majority of the
Numidians swung irreversibly towards a biblically-inspired. form of
Christianity in which the blood-sacrifice of martyrdom replaced tf.-te bloodsacrifice to Saturn. This was central and southern Nurnidia, the heartIand of
the Donatist movement' (ibid. 118). On the ground Donatism was traced
through inscriptions which carried 'recognizably Donatist formulae' and by
association sculptural themes, such as martyrdom (ibid. 69, 231). For Frend
himself (1996,168), one of the duties of archaeology is to release some of those
alternative voices long suppressed by orthodoxy: 'Donatists, Montanists,
Manicheans and Coptic and Nubian Monophysites at last could begin to speak
for themelves through inscriptions, papyri and the steady accumulation of
material evidence'.
A number of Frend's great French North African predecessors had
asserted that the roots of Donatism lay in Berber (i.e. indigenous) nationalism.
A 'Iocal rebuilding' of the incoming doctrine happened again under Islam
when the Maghreb produced a number of sects culminating in that of the
Fatimids which was to spread far and wide. It is certainly tempting to see all
the monumental variations cited here as due to an endemic, indigenous
viewpoint. On this interpretation, the Scots, Picts, Northumbrians and Saxons
would not be expressing their differing politics, but adapting the incoming
The Polifics of Early Medieval Monumentality
Fig. 1.7. Distribution of sculpture in East Yorkshire (Lang 1992): a )
seventh to eighth centuries; b) ninth to eleventh centuries.
doctrines to an unchanging and inevitable individual preference. This is an
ethnically determinist argument which, instead of arguing for immigration of
new ethnic groups to explain cultural.change (e.g. the Anglo-Saxons) attributes
the change to the different reactions of permanently rooted native groups
occupying adjacent territories. For Northumbria at least, this could be
questioned: hers within a territory that extended from York to Lindisfarne,
were produced in the same decade the Lindisfarne Gospels, the Codex
Amiatinus and the Gartort-on-the-Wolds burids (above). This is an era of
experiment rather than adaptation.
Recent scholars have gone out of their way to diminish the differences
between any 'Celtic' church and others (Fletcher 1997, 92; BIair 1996, 6; Hair
and Sharpe 1992). My thesis is intended to bring into focus and interpret such
differences as we can find, and to suggest that behind them lay real conflict.
But it should be noted that I have attempted to explain these differences not
on ethnic or even ~ I t u r agrounds,
l
but on the basis of politics. I have nothing
to say about a 'Celtic' church either, only that during the seventh and eighth
centuries, some of those living in the west appear to have adopted a different
monumental programme to those living in the east, and that this must indicate
something. That something, in my view, would not be an incorrigible ethnicity
but a different way of thinking and a different approach to power. It wouId
probably be sensible at this stage to allow the possibility that deep-rooted local
culture might have an effect on whatever monumentality was practised. But
the facility of communication and awareness of pressure should have allowed
the big issues to be debated in a communal forum. With the variation of
political stance known from North Africa, we can hardly doubt that such
variations were also manifested in the north of Europe, both in the seventh
century and over the next 1,000 years.
References
Alexander, J. G. 1978. Insular Manuscripts, 6th to the 9th Century (A Survey of Manuscripts
illuminated in the British Isles), London: Harvey Wller
Bailey, R. N. 1992. Sutton Hoo and seventh-century art, in R. T. Fame11 and C. Neuman
de Vegvar (eds), SulEon EJoo:$f& years ~~fcer,
31-34, Oxford, Ohio: Miami University
Bede. Historiw Ecclesiastics Gentis Anglomm, B. Colgrave and R. Mynors (eds and trans),
Oxford: Clarendon Press
Blair, J. 1996. Churches in the early English landscape: soda1 and cultural contexts, in J.
Biair and C. Pyrah (eds), Church Archaeology: Research directions jbr the future ( C B A
Res Rep 104), 6-18, York: CBA
Blair, J. and Pyrah, C {eds) 1996. Church Archseology: Research directionsfor thefuture (CBA
Res b p 104), York: CBA
Blair, J. and Sharpe, R. (eds) 1992. Pastoral Care b@re the Parish, Leicester: Leicester
University Press
Bohme, H. W. 1993. Adelsgraber im Frankenreich. Arckologische Zeugrusse zur
Herausbildungeiner Herrenschichtunter dem merowingischenKonigen, Jahrbuchdes
Riirmisch-Gemnischen Zetmlmuseums Mainz, 40, 397534
The Politics of Early Medieval Monumentnlity
21
Bonnet, C. 1993. Les fouilles de I'mcien groupe ipiscopal de Gentve (2976-19931 {Caleers
dfarch6010gie gengvoise l), Geneva
Bradley, R. 1988. Time regained: the creation of continuity, J Brit ArchaeoE Assoc, 140,l-17
Bradley, R. 1993. AIterrng the Earth: the origins of monuments in Britain and continental
Europe: the Rhind Lectures, 1992-92, Edinburgh: Society of Antiquaries of Scotland
Carver, M. 0. H. 1986. Sutton Hoo in context, Settimane di SSfdio, 32, Spoleto: Centro
ftaliano di Studi sd1'Alt.o Medioevo, 77-123
Carver, M. 0. H. 1987. Santa Maria foris Portas at Caste1 Seprio: a famous church in a
new context, World Archaeologj, 18 (3), 312-29
Carver, M. 0. H. 1993. Arguments in Stone: Archaeological research and the European town in
thefirsf nrillennium AD, Oxford: Oxbow Books
Carver, M. 0. H. 1995. Boat burial in Britain: ancient custom or political signal?, in 0.
Crumlin-Pedersen (ed), The Ship as Symbol in Prehistoric and Medieval Scfindimvia,
111-24, Copenhagen: National Museum of Denmark
Carver, M. 0. H. 1998(a). Sutton Hoo. Burial ground of Kings?, London: British Museum
Press
Carver, M. 0.H. 1998@).fherlegungen ziir Bedeutung angel-sachischer Grabhiigel, in
A. Wesse (ed), Studren ziir Archaologie des Ostseeraumes won der Eisenzeit zum
Mittelalter. Festschrfifffiir Michael Miiller-WiEle, 259-68, Neumiinster: Wachholtz
Carver, M. 0.H. 1998(c).Conversion and politics on the eastern seaboard of Britain: some
archaeological indicators, in B. E. Crawford (ed), Conversion and Christianity in the
North Sea World, 1140, St Andrews: University of St Andrews
Carver, M. 0. H. 1999{a). Review of Ian Stead, The Salisbuy Hoard, British Museum
Magazine, (Spring 1999), 39
Carver, M. 0. H. 1999@).Exploring, explaining, imagining: Anglo-Saxon archaeology
1998, in C. Karkov (ed), The Archaeology of Anglo-Saxon England: basic readings, 25-52,
New York and London: Garland
Carver, M. 0.H, in press. Burial as poeby: the case of Sutton Hoo, in E. Tyler (ed), AngloSaxon Treasure, York: Centre for Medieval Studies
Childe, V. G. 1929. The Danube in P~ehistoy,Oxford: Clarendon Press
Chitty, ,Q. J. 1977. The Desert a City: Egiitian and Palestinian monasticism under a Christian
empire
Cramp, R. J. 1980. The Viking image, in R. Farrdl (ed), The Vikings, 8 1 9 , London:
Plullimore ,
Fisher, I. 1996. The west of Scotland, in Blair and Pyrah, 3741
Fletcher, R. 1997. The Conversion of Europe, Berkeley: University of California Press
Frend, W. H. C, 1996. The Archeology of Early Christianihj. A history, London: Chapman
Foot, S. 1990. What was an early Anglo-Saxon monastery?, in J. Loades (ed), Monastic
Studies: the continuity of a tradition, 48-57, Bangor: Headstart History
Geake, H. 1997. f i e Use of Grnve Goods in Cornersion-period England c.600-c.850 ,(Brit
Archaeol Rep 261), Oxford: BAR
Hawkes, J. 1999. Statements in stone. Anglo-Saxon sculpture, Whitby and the
Christianization of the North, in C. Karkov (ed), The Archaeology of Anglo-Smon
England: basic readings, 403-21, New York and London: Garland
Henderson, G, 1999. Vision and Image in Early Christian England, Cambridge: Cambridge
University Press
Hill, P. 1997. Whithorn and St Ninian: the excavation of a monastic town 1984-1991, Whithom
and Stroud: Whithorn Trust and Sutt-onPublishing
22
Marfin Carver
Hines, J. 1984. The Scandinavian Character of AngIian England in the Pre-Viking Period (Brit
Archaeol Rep 1241, Oxford: BAR
J o h n , S. 1983. Burgh Castle: Excavatrons by Charles Green 1958-61 (East AngIian Archaeol
Rep 20), Gressenhall: Norfolk Museum Service
Lang, 1. 1991. Corpus of Anglo-Saxon Sculpture, 3, York ~ n dYorkshire, Oxford: British
Academy / Oxford University Press
Levison, W. 1946. England and the°Continent in the Eighth Century, Oxford: Oxford
University Press
Lowe, C, forthcoming. Exmvntions at Hoddom, Dumfriesshire: An wlclosed holy place of the late
first lnilIennium AD, Edinburgh: Historic Scotland
Lutovsky, M. 1996. Between Sutton Hoo and Chemaya Mogla: barrows in eastern and
western early medieval Europe, Antiquity, 70, 6726
Mango, M. M,, Mango, C., Evans, A. and Hughes, M. 1989, A 6th century Mediterranean
bucket from Bromeswell parish, Suffolk, Antiquify, 63, 295-311
Mitchell, P. and Ryan, M. 1998. Reading the Irish Landscape, Dublin: Town House
Miiller-Wille, M. 1998(a). Zwei reliogiose Welten: Bestattungen der frankischen Konige
Childerich und Chlodwig, Abhmdlung dder Akadewaie dur Wissenschfien und der
Liternfur, Mainz: Steiner Verlag
Miiller-Wille, M. 2998(b). The Cross as a symbol of personal Christian belief in a changing
religious world. Examples from selected areas in Merovingian and Carolingian
Europe, KVHAA Konferenser 40, -179-200, Stockholm: KVHAA
Myturn, H. 1992. The Origins of Early Christian Ireland, London: RoutIedge
Opie, I. and Opie, P. 1977. The Lore and language of School Children, London: Granada
Perinetti, R. 1989. Augusta Praetoria. Le necropoIi christiane, Acfes du XI Coragres
Infernational d'archiologae Chritienne, 121526, Vatican City
Proudfoot, E. 1996. Excavations at the long-cist cemetery on the Hallow Hill, St Andrews,
Fife 1975-7, Pruc Soc Antiq Scot, 126,387-454
Proudfoot, E. 1998. The Hauowhill and the origins of Christianity in Scotland, in B. E.
Crawford (ed), Co~vmsionand Christianity in the North Sea World, 57-74, St Andrews:
University of St Andrews
Ramquist, P. H. 1992. Hiigom. The excavations 1949-1984, Urn&: University of Urn&
Ritchie, A. 1995. Meigle and lay patronage in the 9th and 10th centuries AD, Tayside and
Fife Archaeological Journal, 1, 1-10
Sharpe, R. (trans) 1995. Adomnnn of Iona. Life of St Columba, H ~ n d s w o r t h Penguin
:
Shephard, J. 1979. The social identity of the individual in isolated barrows and barrow
cemeteries in Anglo-Saxon England, in B. Burnham and J. Kingsbury (eds), Space,
Hierarchy nnd Society: interdisciplinay studies in social area analysis (Brit Archaeol Rep,
Int Ser, 59), 47-79, Oxford: BAR
Smith, I. 1996. The archaeology of the early Christian church in Scotland and Man AD
400-1200, in Blair and Pyrah, 19-37
Slupecki, L. P. 1997. Einfliisse des Christenhurts auf die heidnische Religion der
Ostseeslawen irn 8.-12. Jahrhundert: Tempel - Gottbilder - Kult, in M. Miiller-Wille
(ed), Rom und Byzanz im Norden. Mission und Glazdbenswechsel iwr Ostseeraurn wlihrend
des E-14. Jahrhzdnderts, 177-89, Manz: Akademie der Wissenschaften und der
Litera tur
Taylor, H. M. and Taylor, J. 1965. Anglo-Saxon Architecture, Cambridge: Cambridge
University Press
van de Noort, R. 1993. The context of early medieval barrows in Western Europe,
Antiquity, 67, 66-73